It is unknown whether growth differentiation factor 15 (GDF-15) is associated with chronic musculoskeletal pain (CMP) and whether or not its association with incident cardiovascular disease (CVD) changes according to CMP status. 1,957 randomly-selected adults aged ≥65 years without prior CVD were followed up between 2015-2023. CMP was classified according to its intensity, frequency, and interference with daily activities. The association between GDF-15 levels and CMP was assessed using linear models with progressive inclusion of potential confounders, whereas the association between GDF-15 and CVD risk was evaluated with Cox-proportional hazard models with similar adjustment and interaction terms between GDF-15 and CMP. The incremental predictive performance of GDF-15 over standard predictors was evaluated using discrimination and risk reclassification metrics. GDF-15 concentrations were 6.90% (95%CI:2.56;11.25) higher in individuals with CMP, and up to 8.89% (4.07;15.71) and 15.79% (8.43;23.16) higher in those with ≥3 CMP locations and interfering pain. These increased levels were influenced by a higher prevalence of cardiometabolic risk factors, functional impairments, depressive symptoms, and greater levels of inflammation in individuals with CMP. In fully-adjusted models, a two-fold increase in GDF-15 was associated with a with a 1.49 increased risk (95%CI: 1.08; 2.05) of a CVD event in individuals with CMP, but not among those without CMP [1.02 (0.77; 1.35)]; p-interaction 0.041. Adding GDF-15 to models including the Framingham Risk Score improved predictive performance among individuals with CMP. We provide evidence that GDF-15 could serve as a biomarker to assess CMP, as well as to predict CVD incidence in individuals with CMP.